diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000000000000000000000000000000000000..a6344aac8c09253b3b630fb776ae94478aa0275b --- /dev/null +++ b/.gitattributes @@ -0,0 +1,35 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text diff --git a/checkpoints/012000/pretrained_model/config.json b/checkpoints/012000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..55c4c482785119428129895740682e4147a64b12 --- /dev/null +++ b/checkpoints/012000/pretrained_model/config.json @@ -0,0 +1,87 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 +} \ No newline at end of file diff --git a/checkpoints/012000/pretrained_model/model.safetensors b/checkpoints/012000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..dbabfdbf3f5aceb43c374b487d82e4ded872e343 --- /dev/null +++ b/checkpoints/012000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:906f4850e710d8e50e3c22a445f191666780a199941166875ed60a0fda962dc8 +size 1051882616 diff --git a/checkpoints/012000/pretrained_model/policy_postprocessor.json b/checkpoints/012000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..f2c5e6cea7850bb999ab193f96021fa0129ce4db --- /dev/null +++ b/checkpoints/012000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/checkpoints/012000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/012000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/012000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/012000/pretrained_model/policy_preprocessor.json b/checkpoints/012000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..b998da36fac4f946708f548d498c10aba186f303 --- /dev/null +++ b/checkpoints/012000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,56 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/012000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/012000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/012000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/012000/pretrained_model/train_config.json b/checkpoints/012000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..4770c83d303b69a8ebba87255335138e2c940b57 --- /dev/null +++ b/checkpoints/012000/pretrained_model/train_config.json @@ -0,0 +1,223 @@ +{ + "dataset": { + "repo_id": "jjr1007/may7_merged", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "return_uint8": false, + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 + }, + "reward_model": null, + "output_dir": "outputs/run_diffusion", + "job_name": "diffusion", + "resume": false, + "seed": 1000, + "cudnn_deterministic": false, + "num_workers": 4, + "batch_size": 90, + "prefetch_factor": 4, + "persistent_workers": true, + "steps": 30000, + "eval_freq": 1000, + "log_freq": 100, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 2000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 19, + "use_async_envs": true + }, + "wandb": { + "enable": true, + "disable_artifact": false, + "project": "multitask-dit-experiments", + "entity": null, + "notes": null, + "run_id": "diffusion_ddpm_h64a32", + "mode": null, + "add_tags": true + }, + "peft": null, + "sample_weighting": null, + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/checkpoints/012000/training_state/optimizer_param_groups.json b/checkpoints/012000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..3bab2422eabe2262a591012a92a67874c87c38c8 --- /dev/null +++ b/checkpoints/012000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 6.696194330590151e-05, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/checkpoints/012000/training_state/optimizer_state.safetensors b/checkpoints/012000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..143a17def62be541a568d693db56da1fad6bb8ac --- /dev/null +++ b/checkpoints/012000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:50d2078ce88b242eeafb69774150fb7a82c57f6ae498649af559ce27f4b94376 +size 2103678952 diff --git a/checkpoints/012000/training_state/rng_state.safetensors b/checkpoints/012000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..ed976857dba4d35aa58efd888ea21ee4b0a847e0 --- /dev/null +++ b/checkpoints/012000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a38acc9ad9bbbbe253aa0fa4ed4dd4aea3d4a7b4f6948493146f67a88de34e24 +size 15708 diff --git a/checkpoints/012000/training_state/scheduler_state.json b/checkpoints/012000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..28b74e1a7bf54424fa23241c9e34640b6735952c --- /dev/null +++ b/checkpoints/012000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 12000, + "_step_count": 12001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 6.696194330590151e-05 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/checkpoints/012000/training_state/training_step.json b/checkpoints/012000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..c4fb27ac819b81943e6545c7c18510bdfb8eae1b --- /dev/null +++ b/checkpoints/012000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 12000 +} \ No newline at end of file diff --git a/checkpoints/014000/pretrained_model/config.json b/checkpoints/014000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..55c4c482785119428129895740682e4147a64b12 --- /dev/null +++ b/checkpoints/014000/pretrained_model/config.json @@ -0,0 +1,87 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 +} \ No newline at end of file diff --git a/checkpoints/014000/pretrained_model/model.safetensors b/checkpoints/014000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..1353e18e01d8b12c8a639e56f2fd3f4170afaa05 --- /dev/null +++ b/checkpoints/014000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:06723cc4738fe032b93237d129558aeb7f3b7679d86515345467ec5c7c894a90 +size 1051882616 diff --git a/checkpoints/014000/pretrained_model/policy_postprocessor.json b/checkpoints/014000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..f2c5e6cea7850bb999ab193f96021fa0129ce4db --- /dev/null +++ b/checkpoints/014000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/checkpoints/014000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/014000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/014000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/014000/pretrained_model/policy_preprocessor.json b/checkpoints/014000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..b998da36fac4f946708f548d498c10aba186f303 --- /dev/null +++ b/checkpoints/014000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,56 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/014000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/014000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/014000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/014000/pretrained_model/train_config.json b/checkpoints/014000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..4770c83d303b69a8ebba87255335138e2c940b57 --- /dev/null +++ b/checkpoints/014000/pretrained_model/train_config.json @@ -0,0 +1,223 @@ +{ + "dataset": { + "repo_id": "jjr1007/may7_merged", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "return_uint8": false, + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 + }, + "reward_model": null, + "output_dir": "outputs/run_diffusion", + "job_name": "diffusion", + "resume": false, + "seed": 1000, + "cudnn_deterministic": false, + "num_workers": 4, + "batch_size": 90, + "prefetch_factor": 4, + "persistent_workers": true, + "steps": 30000, + "eval_freq": 1000, + "log_freq": 100, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 2000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 19, + "use_async_envs": true + }, + "wandb": { + "enable": true, + "disable_artifact": false, + "project": "multitask-dit-experiments", + "entity": null, + "notes": null, + "run_id": "diffusion_ddpm_h64a32", + "mode": null, + "add_tags": true + }, + "peft": null, + "sample_weighting": null, + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/checkpoints/014000/training_state/optimizer_param_groups.json b/checkpoints/014000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..4aebe39cccd2e2ec2eb2e396cfa4fe6919163413 --- /dev/null +++ b/checkpoints/014000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 5.6636276364186105e-05, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/checkpoints/014000/training_state/optimizer_state.safetensors b/checkpoints/014000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..3ea4b3c6dad343f16f318cf0474f7c8f8f7c0086 --- /dev/null +++ b/checkpoints/014000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc38527160de33876c1b6c3a97dcde2e62a4fd47073c6a2fb6610f60ba7848dc +size 2103678952 diff --git a/checkpoints/014000/training_state/rng_state.safetensors b/checkpoints/014000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..eaa095469c14faaf6dd6001d81d1e96c47f0251f --- /dev/null +++ b/checkpoints/014000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cbe30a54abfb98215df046c87f0667ee3b5b4bb4748cc7c5e96d1a67884b144 +size 15708 diff --git a/checkpoints/014000/training_state/scheduler_state.json b/checkpoints/014000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..8131d8499bdd543c028782a0787510d92627c3eb --- /dev/null +++ b/checkpoints/014000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 14000, + "_step_count": 14001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 5.6636276364186105e-05 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/checkpoints/014000/training_state/training_step.json b/checkpoints/014000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..560b82990f5c806d547ed82a5432ba60aec285e1 --- /dev/null +++ b/checkpoints/014000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 14000 +} \ No newline at end of file diff --git a/checkpoints/016000/pretrained_model/config.json b/checkpoints/016000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..55c4c482785119428129895740682e4147a64b12 --- /dev/null +++ b/checkpoints/016000/pretrained_model/config.json @@ -0,0 +1,87 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 +} \ No newline at end of file diff --git a/checkpoints/016000/pretrained_model/model.safetensors b/checkpoints/016000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e94c54fb5696207eaabeba51eb8162aa2dde4977 --- /dev/null +++ b/checkpoints/016000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2b86e534598f8c42b5e2df35f7f40deb69d6e6c26c06ad2aa3c31b91d22d59f5 +size 1051882616 diff --git a/checkpoints/016000/pretrained_model/policy_postprocessor.json b/checkpoints/016000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..f2c5e6cea7850bb999ab193f96021fa0129ce4db --- /dev/null +++ b/checkpoints/016000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/checkpoints/016000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/016000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/016000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/016000/pretrained_model/policy_preprocessor.json b/checkpoints/016000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..b998da36fac4f946708f548d498c10aba186f303 --- /dev/null +++ b/checkpoints/016000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,56 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/016000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/016000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/016000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/016000/pretrained_model/train_config.json b/checkpoints/016000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..4770c83d303b69a8ebba87255335138e2c940b57 --- /dev/null +++ b/checkpoints/016000/pretrained_model/train_config.json @@ -0,0 +1,223 @@ +{ + "dataset": { + "repo_id": "jjr1007/may7_merged", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "return_uint8": false, + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 + }, + "reward_model": null, + "output_dir": "outputs/run_diffusion", + "job_name": "diffusion", + "resume": false, + "seed": 1000, + "cudnn_deterministic": false, + "num_workers": 4, + "batch_size": 90, + "prefetch_factor": 4, + "persistent_workers": true, + "steps": 30000, + "eval_freq": 1000, + "log_freq": 100, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 2000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 19, + "use_async_envs": true + }, + "wandb": { + "enable": true, + "disable_artifact": false, + "project": "multitask-dit-experiments", + "entity": null, + "notes": null, + "run_id": "diffusion_ddpm_h64a32", + "mode": null, + "add_tags": true + }, + "peft": null, + "sample_weighting": null, + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/checkpoints/016000/training_state/optimizer_param_groups.json b/checkpoints/016000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..ceb0e9be6e0b0b0752ccc8dfdebcd7e06df5e1ae --- /dev/null +++ b/checkpoints/016000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 4.601069472234584e-05, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/checkpoints/016000/training_state/optimizer_state.safetensors b/checkpoints/016000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b3c2a8d519748693b026d25e115dc425c84129ea --- /dev/null +++ b/checkpoints/016000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c13bd67448cfa41044df8ffb4ff493f37cac5fdecb89309f843f0e55f597496b +size 2103678952 diff --git a/checkpoints/016000/training_state/rng_state.safetensors b/checkpoints/016000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..4aa7fcb9a27e58dabc9c6a5ee53a9d6c762e20b5 --- /dev/null +++ b/checkpoints/016000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d9bf79afcc96f31d7a82fcc8dc7f28f94500aa644dc9a9117ae76ee02644ba42 +size 15708 diff --git a/checkpoints/016000/training_state/scheduler_state.json b/checkpoints/016000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..9ee3d3ccb404d8a2b59f5b15caf449f22566d041 --- /dev/null +++ b/checkpoints/016000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 16000, + "_step_count": 16001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 4.601069472234584e-05 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/checkpoints/016000/training_state/training_step.json b/checkpoints/016000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..798020d90d4c39414dfab88708ecefd52f2e9ac5 --- /dev/null +++ b/checkpoints/016000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 16000 +} \ No newline at end of file diff --git a/checkpoints/018000/pretrained_model/config.json b/checkpoints/018000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..55c4c482785119428129895740682e4147a64b12 --- /dev/null +++ b/checkpoints/018000/pretrained_model/config.json @@ -0,0 +1,87 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 +} \ No newline at end of file diff --git a/checkpoints/018000/pretrained_model/model.safetensors b/checkpoints/018000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..bd508939d84aae2100bc6467e68d51a7fef93855 --- /dev/null +++ b/checkpoints/018000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:46ef138cacd355a084adaeebcdcc15413da9a92e4efa94445d880019f9265c26 +size 1051882616 diff --git a/checkpoints/018000/pretrained_model/policy_postprocessor.json b/checkpoints/018000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..f2c5e6cea7850bb999ab193f96021fa0129ce4db --- /dev/null +++ b/checkpoints/018000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/checkpoints/018000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/018000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/018000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/018000/pretrained_model/policy_preprocessor.json b/checkpoints/018000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..b998da36fac4f946708f548d498c10aba186f303 --- /dev/null +++ b/checkpoints/018000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,56 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/018000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/018000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/018000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/018000/pretrained_model/train_config.json b/checkpoints/018000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..4770c83d303b69a8ebba87255335138e2c940b57 --- /dev/null +++ b/checkpoints/018000/pretrained_model/train_config.json @@ -0,0 +1,223 @@ +{ + "dataset": { + "repo_id": "jjr1007/may7_merged", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "return_uint8": false, + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 + }, + "reward_model": null, + "output_dir": "outputs/run_diffusion", + "job_name": "diffusion", + "resume": false, + "seed": 1000, + "cudnn_deterministic": false, + "num_workers": 4, + "batch_size": 90, + "prefetch_factor": 4, + "persistent_workers": true, + "steps": 30000, + "eval_freq": 1000, + "log_freq": 100, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 2000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 19, + "use_async_envs": true + }, + "wandb": { + "enable": true, + "disable_artifact": false, + "project": "multitask-dit-experiments", + "entity": null, + "notes": null, + "run_id": "diffusion_ddpm_h64a32", + "mode": null, + "add_tags": true + }, + "peft": null, + "sample_weighting": null, + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/checkpoints/018000/training_state/optimizer_param_groups.json b/checkpoints/018000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..92fce0a675e6976e9078aa2472c5b21e8766aaf6 --- /dev/null +++ b/checkpoints/018000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 3.556540263301896e-05, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/checkpoints/018000/training_state/optimizer_state.safetensors b/checkpoints/018000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..240e118f2069806e0f4b38a4eb205f912825dbd6 --- /dev/null +++ b/checkpoints/018000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5c9847ae0857325d9be42d724f95bbc273e39714eabc6e15caaddf53e9a2342e +size 2103678952 diff --git a/checkpoints/018000/training_state/rng_state.safetensors b/checkpoints/018000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b079731269b009b08faf1350ecaa30757365c2a0 --- /dev/null +++ b/checkpoints/018000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:15096cbbc9a12bf9d4ce7cac17eb04703ca0f0075a49b16fdef8201cd85cfe70 +size 15708 diff --git a/checkpoints/018000/training_state/scheduler_state.json b/checkpoints/018000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..8eda9a123424f8bca5069100baddfdf40bfb2206 --- /dev/null +++ b/checkpoints/018000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 18000, + "_step_count": 18001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 3.556540263301896e-05 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/checkpoints/018000/training_state/training_step.json b/checkpoints/018000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..332a3c17a399c305da7a515c33ffb382d0aff339 --- /dev/null +++ b/checkpoints/018000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 18000 +} \ No newline at end of file diff --git a/checkpoints/020000/pretrained_model/config.json b/checkpoints/020000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..55c4c482785119428129895740682e4147a64b12 --- /dev/null +++ b/checkpoints/020000/pretrained_model/config.json @@ -0,0 +1,87 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 +} \ No newline at end of file diff --git a/checkpoints/020000/pretrained_model/model.safetensors b/checkpoints/020000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..517f4a63876715abd09aa35e1b9668e7140b7921 --- /dev/null +++ b/checkpoints/020000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5620f82ca7fef1e77964ec9292d02f2d155da2f4e2eab6a88d4f4a3d054ced54 +size 1051882616 diff --git a/checkpoints/020000/pretrained_model/policy_postprocessor.json b/checkpoints/020000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..f2c5e6cea7850bb999ab193f96021fa0129ce4db --- /dev/null +++ b/checkpoints/020000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/checkpoints/020000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/020000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/020000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/020000/pretrained_model/policy_preprocessor.json b/checkpoints/020000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..b998da36fac4f946708f548d498c10aba186f303 --- /dev/null +++ b/checkpoints/020000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,56 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/020000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/020000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/020000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/020000/pretrained_model/train_config.json b/checkpoints/020000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..4770c83d303b69a8ebba87255335138e2c940b57 --- /dev/null +++ b/checkpoints/020000/pretrained_model/train_config.json @@ -0,0 +1,223 @@ +{ + "dataset": { + "repo_id": "jjr1007/may7_merged", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "return_uint8": false, + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 + }, + "reward_model": null, + "output_dir": "outputs/run_diffusion", + "job_name": "diffusion", + "resume": false, + "seed": 1000, + "cudnn_deterministic": false, + "num_workers": 4, + "batch_size": 90, + "prefetch_factor": 4, + "persistent_workers": true, + "steps": 30000, + "eval_freq": 1000, + "log_freq": 100, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 2000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 19, + "use_async_envs": true + }, + "wandb": { + "enable": true, + "disable_artifact": false, + "project": "multitask-dit-experiments", + "entity": null, + "notes": null, + "run_id": "diffusion_ddpm_h64a32", + "mode": null, + "add_tags": true + }, + "peft": null, + "sample_weighting": null, + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/checkpoints/020000/training_state/optimizer_param_groups.json b/checkpoints/020000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..3ee7181f17c0d6dea7426c9f634454c2643c67d0 --- /dev/null +++ b/checkpoints/020000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 2.5772456483367497e-05, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/checkpoints/020000/training_state/optimizer_state.safetensors b/checkpoints/020000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8ae1d366a7720e6ca72b28064e8977a3c30c8d1e --- /dev/null +++ b/checkpoints/020000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a7e5b56e581dbe3e517ad5263c8be97f279d6a3a42c68108f50494fc4faa0f5c +size 2103678952 diff --git a/checkpoints/020000/training_state/rng_state.safetensors b/checkpoints/020000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c3fc64afa8f8ee790d80bf99ad79b6e1366ab6cf --- /dev/null +++ b/checkpoints/020000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:478242bd225b829275c229099d7f0b37c85f10fc0b4ced539fdba51bb7d3a654 +size 15708 diff --git a/checkpoints/020000/training_state/scheduler_state.json b/checkpoints/020000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..e64a8f7b5e0b2f21976708be47810bc09371c7ac --- /dev/null +++ b/checkpoints/020000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 20000, + "_step_count": 20001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 2.5772456483367497e-05 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/checkpoints/020000/training_state/training_step.json b/checkpoints/020000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..dc9bb47026c5d5237ca6fc5dbff6020dd122ea05 --- /dev/null +++ b/checkpoints/020000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 20000 +} \ No newline at end of file diff --git a/checkpoints/022000/pretrained_model/config.json b/checkpoints/022000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..55c4c482785119428129895740682e4147a64b12 --- /dev/null +++ b/checkpoints/022000/pretrained_model/config.json @@ -0,0 +1,87 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 +} \ No newline at end of file diff --git a/checkpoints/022000/pretrained_model/model.safetensors b/checkpoints/022000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..242582d6f22b3ccc6921f444790bda3b01039781 --- /dev/null +++ b/checkpoints/022000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8f19e377db9cbdfa160274ead19d6c6295de686060c6eea8fd9892a9019f45a1 +size 1051882616 diff --git a/checkpoints/022000/pretrained_model/policy_postprocessor.json b/checkpoints/022000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..f2c5e6cea7850bb999ab193f96021fa0129ce4db --- /dev/null +++ b/checkpoints/022000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/checkpoints/022000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/022000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/022000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/022000/pretrained_model/policy_preprocessor.json b/checkpoints/022000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..b998da36fac4f946708f548d498c10aba186f303 --- /dev/null +++ b/checkpoints/022000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,56 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/022000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/022000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/022000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/022000/pretrained_model/train_config.json b/checkpoints/022000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..4770c83d303b69a8ebba87255335138e2c940b57 --- /dev/null +++ b/checkpoints/022000/pretrained_model/train_config.json @@ -0,0 +1,223 @@ +{ + "dataset": { + "repo_id": "jjr1007/may7_merged", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "return_uint8": false, + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 + }, + "reward_model": null, + "output_dir": "outputs/run_diffusion", + "job_name": "diffusion", + "resume": false, + "seed": 1000, + "cudnn_deterministic": false, + "num_workers": 4, + "batch_size": 90, + "prefetch_factor": 4, + "persistent_workers": true, + "steps": 30000, + "eval_freq": 1000, + "log_freq": 100, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 2000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 19, + "use_async_envs": true + }, + "wandb": { + "enable": true, + "disable_artifact": false, + "project": "multitask-dit-experiments", + "entity": null, + "notes": null, + "run_id": "diffusion_ddpm_h64a32", + "mode": null, + "add_tags": true + }, + "peft": null, + "sample_weighting": null, + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/checkpoints/022000/training_state/optimizer_param_groups.json b/checkpoints/022000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..fddafb062aae6c24bee981e605b0ecde955057ed --- /dev/null +++ b/checkpoints/022000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 1.7074431046748075e-05, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/checkpoints/022000/training_state/optimizer_state.safetensors b/checkpoints/022000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..6fbef9ecbd9a9b5f4328da1e6fbcda73cfcf355d --- /dev/null +++ b/checkpoints/022000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b932b364d5beddbb59801c34f0fa70e590463bab11a578ff3af53cdbb8e18e9a +size 2103678952 diff --git a/checkpoints/022000/training_state/rng_state.safetensors b/checkpoints/022000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..826cddd055aa1b014b800e71a8934b816e6c4e47 --- /dev/null +++ b/checkpoints/022000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5c0ae791aae43cbc968677aca4e23ee321e2dc473e69db607afbbf311a004ed2 +size 15708 diff --git a/checkpoints/022000/training_state/scheduler_state.json b/checkpoints/022000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..838004d5026303f1598c395d750648890c7ea99a --- /dev/null +++ b/checkpoints/022000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 22000, + "_step_count": 22001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 1.7074431046748075e-05 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/checkpoints/022000/training_state/training_step.json b/checkpoints/022000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..d60e27dffb5d802ca16598a852aec850971cbb39 --- /dev/null +++ b/checkpoints/022000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 22000 +} \ No newline at end of file diff --git a/checkpoints/024000/pretrained_model/config.json b/checkpoints/024000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..55c4c482785119428129895740682e4147a64b12 --- /dev/null +++ b/checkpoints/024000/pretrained_model/config.json @@ -0,0 +1,87 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 +} \ No newline at end of file diff --git a/checkpoints/024000/pretrained_model/model.safetensors b/checkpoints/024000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..979bf086ce72e62bf25dd80ef6868485fd93ea71 --- /dev/null +++ b/checkpoints/024000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:108ad4665e582e5e22fede7c3e59099bf838045b8c4feb513a8b598a74c464a5 +size 1051882616 diff --git a/checkpoints/024000/pretrained_model/policy_postprocessor.json b/checkpoints/024000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..f2c5e6cea7850bb999ab193f96021fa0129ce4db --- /dev/null +++ b/checkpoints/024000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/checkpoints/024000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/024000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/024000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/024000/pretrained_model/policy_preprocessor.json b/checkpoints/024000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..b998da36fac4f946708f548d498c10aba186f303 --- /dev/null +++ b/checkpoints/024000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,56 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/024000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/024000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/024000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/024000/pretrained_model/train_config.json b/checkpoints/024000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..4770c83d303b69a8ebba87255335138e2c940b57 --- /dev/null +++ b/checkpoints/024000/pretrained_model/train_config.json @@ -0,0 +1,223 @@ +{ + "dataset": { + "repo_id": "jjr1007/may7_merged", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "return_uint8": false, + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 + }, + "reward_model": null, + "output_dir": "outputs/run_diffusion", + "job_name": "diffusion", + "resume": false, + "seed": 1000, + "cudnn_deterministic": false, + "num_workers": 4, + "batch_size": 90, + "prefetch_factor": 4, + "persistent_workers": true, + "steps": 30000, + "eval_freq": 1000, + "log_freq": 100, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 2000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 19, + "use_async_envs": true + }, + "wandb": { + "enable": true, + "disable_artifact": false, + "project": "multitask-dit-experiments", + "entity": null, + "notes": null, + "run_id": "diffusion_ddpm_h64a32", + "mode": null, + "add_tags": true + }, + "peft": null, + "sample_weighting": null, + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/checkpoints/024000/training_state/optimizer_param_groups.json b/checkpoints/024000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..a05f3bcc4c84b56f021e7c83365181a158e4c4f6 --- /dev/null +++ b/checkpoints/024000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 9.864418103451828e-06, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/checkpoints/024000/training_state/optimizer_state.safetensors b/checkpoints/024000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..56f528f530be44a918d5cfc1f6236130480f1d47 --- /dev/null +++ b/checkpoints/024000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:14f85426a94d4a356f3f76db5bf1c4dc1fda7c66a9194afc3733ad127178c869 +size 2103678952 diff --git a/checkpoints/024000/training_state/rng_state.safetensors b/checkpoints/024000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..9e28980e29ad74a31bfde768a525a9b7c034d27d --- /dev/null +++ b/checkpoints/024000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c62654015db3a88dbafa10aa315b2ffddc1a47fae29058e275301c748f7e2a81 +size 15708 diff --git a/checkpoints/024000/training_state/scheduler_state.json b/checkpoints/024000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..712a2964d03d13352419cfe070e750e03b63e265 --- /dev/null +++ b/checkpoints/024000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 24000, + "_step_count": 24001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 9.864418103451828e-06 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/checkpoints/024000/training_state/training_step.json b/checkpoints/024000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..4bf22a00d4c8b45342d351916c6e2f27ab1dd7d9 --- /dev/null +++ b/checkpoints/024000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 24000 +} \ No newline at end of file diff --git a/checkpoints/026000/pretrained_model/config.json b/checkpoints/026000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..55c4c482785119428129895740682e4147a64b12 --- /dev/null +++ b/checkpoints/026000/pretrained_model/config.json @@ -0,0 +1,87 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 +} \ No newline at end of file diff --git a/checkpoints/026000/pretrained_model/model.safetensors b/checkpoints/026000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..39e843eedc3d8fd8821df66ec47399ccd35811d4 --- /dev/null +++ b/checkpoints/026000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:64c68a97f3ef85f6b0785691e998fee04b9c8ca8773d5d1f701a848b4380c1fc +size 1051882616 diff --git a/checkpoints/026000/pretrained_model/policy_postprocessor.json b/checkpoints/026000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..f2c5e6cea7850bb999ab193f96021fa0129ce4db --- /dev/null +++ b/checkpoints/026000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/checkpoints/026000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/026000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/026000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/026000/pretrained_model/policy_preprocessor.json b/checkpoints/026000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..b998da36fac4f946708f548d498c10aba186f303 --- /dev/null +++ b/checkpoints/026000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,56 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/026000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/026000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/026000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/026000/pretrained_model/train_config.json b/checkpoints/026000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..4770c83d303b69a8ebba87255335138e2c940b57 --- /dev/null +++ b/checkpoints/026000/pretrained_model/train_config.json @@ -0,0 +1,223 @@ +{ + "dataset": { + "repo_id": "jjr1007/may7_merged", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "return_uint8": false, + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 + }, + "reward_model": null, + "output_dir": "outputs/run_diffusion", + "job_name": "diffusion", + "resume": false, + "seed": 1000, + "cudnn_deterministic": false, + "num_workers": 4, + "batch_size": 90, + "prefetch_factor": 4, + "persistent_workers": true, + "steps": 30000, + "eval_freq": 1000, + "log_freq": 100, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 2000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 19, + "use_async_envs": true + }, + "wandb": { + "enable": true, + "disable_artifact": false, + "project": "multitask-dit-experiments", + "entity": null, + "notes": null, + "run_id": "diffusion_ddpm_h64a32", + "mode": null, + "add_tags": true + }, + "peft": null, + "sample_weighting": null, + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/checkpoints/026000/training_state/optimizer_param_groups.json b/checkpoints/026000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..799037be7b540d098a3aab15cb9bacec9974825a --- /dev/null +++ b/checkpoints/026000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 4.468261357254339e-06, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/checkpoints/026000/training_state/optimizer_state.safetensors b/checkpoints/026000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0d126695568fa760edb1c0bacdee4cbff9079ecf --- /dev/null +++ b/checkpoints/026000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:49d9c058bd182e6d20b8db34503a8a6df4c6f014420432224437c14de951e7df +size 2103678952 diff --git a/checkpoints/026000/training_state/rng_state.safetensors b/checkpoints/026000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..592ea2bd3658dae994bd93e54b3d457a02baff86 --- /dev/null +++ b/checkpoints/026000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:64072b48a8be48cfdbafb1f59c89a6f82b97613134962e0dabb431b64a919de7 +size 15708 diff --git a/checkpoints/026000/training_state/scheduler_state.json b/checkpoints/026000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..cbfe65b8e981e607254f5d7d5e74e098c3fb81dd --- /dev/null +++ b/checkpoints/026000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 26000, + "_step_count": 26001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 4.468261357254339e-06 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/checkpoints/026000/training_state/training_step.json b/checkpoints/026000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..60495b67a6a5c8ae0694d967a703afdf4bdc4066 --- /dev/null +++ b/checkpoints/026000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 26000 +} \ No newline at end of file diff --git a/checkpoints/028000/pretrained_model/config.json b/checkpoints/028000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..55c4c482785119428129895740682e4147a64b12 --- /dev/null +++ b/checkpoints/028000/pretrained_model/config.json @@ -0,0 +1,87 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 +} \ No newline at end of file diff --git a/checkpoints/028000/pretrained_model/model.safetensors b/checkpoints/028000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..97a3f93a0cd8746500db52a9f06a1361508413fc --- /dev/null +++ b/checkpoints/028000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7d4041bef7411f4e35b1473f7f78f650e574471d5482bf8ba5c5a6b80ac1103e +size 1051882616 diff --git a/checkpoints/028000/pretrained_model/policy_postprocessor.json b/checkpoints/028000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..f2c5e6cea7850bb999ab193f96021fa0129ce4db --- /dev/null +++ b/checkpoints/028000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/checkpoints/028000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/028000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/028000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/028000/pretrained_model/policy_preprocessor.json b/checkpoints/028000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..b998da36fac4f946708f548d498c10aba186f303 --- /dev/null +++ b/checkpoints/028000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,56 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/028000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/028000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..15249c62daa3666457cdf575d4b537acbf21ae4d --- /dev/null +++ b/checkpoints/028000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf7ca29660235d29e109d86effbaa839957cb210ab49d9084e38e90228d4f11 +size 6560 diff --git a/checkpoints/028000/pretrained_model/train_config.json b/checkpoints/028000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..4770c83d303b69a8ebba87255335138e2c940b57 --- /dev/null +++ b/checkpoints/028000/pretrained_model/train_config.json @@ -0,0 +1,223 @@ +{ + "dataset": { + "repo_id": "jjr1007/may7_merged", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "return_uint8": false, + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 1080, + 1920 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "jjr1007/diffusion_no_dit", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 64, + "n_action_steps": 32, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": [ + 320, + 240 + ], + "crop_ratio": 1.0, + "crop_shape": null, + "crop_is_random": true, + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "use_group_norm": false, + "spatial_softmax_num_keypoints": 32, + "use_separate_rgb_encoder_per_camera": true, + "down_dims": [ + 512, + 1024, + 2048 + ], + "kernel_size": 5, + "n_groups": 8, + "diffusion_step_embed_dim": 128, + "use_film_scale_modulation": true, + "noise_scheduler_type": "DDPM", + "num_train_timesteps": 100, + "beta_schedule": "squaredcos_cap_v2", + "beta_start": 0.0001, + "beta_end": 0.02, + "prediction_type": "epsilon", + "clip_sample": true, + "clip_sample_range": 1.0, + "num_inference_steps": null, + "compile_model": false, + "compile_mode": "reduce-overhead", + "do_mask_loss_for_padding": false, + "optimizer_lr": 0.0001, + "optimizer_betas": [ + 0.95, + 0.999 + ], + "optimizer_eps": 1e-08, + "optimizer_weight_decay": 1e-06, + "scheduler_name": "cosine", + "scheduler_warmup_steps": 500 + }, + "reward_model": null, + "output_dir": "outputs/run_diffusion", + "job_name": "diffusion", + "resume": false, + "seed": 1000, + "cudnn_deterministic": false, + "num_workers": 4, + "batch_size": 90, + "prefetch_factor": 4, + "persistent_workers": true, + "steps": 30000, + "eval_freq": 1000, + "log_freq": 100, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 2000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 19, + "use_async_envs": true + }, + "wandb": { + "enable": true, + "disable_artifact": false, + "project": "multitask-dit-experiments", + "entity": null, + "notes": null, + "run_id": "diffusion_ddpm_h64a32", + "mode": null, + "add_tags": true + }, + "peft": null, + "sample_weighting": null, + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/checkpoints/028000/training_state/optimizer_param_groups.json b/checkpoints/028000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..3f74f7d2f50834f703ae0b0c6550a17d34fcb66d --- /dev/null +++ b/checkpoints/028000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 1.1298305091066664e-06, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/checkpoints/028000/training_state/optimizer_state.safetensors b/checkpoints/028000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..98fc1191a8e529a1a58587bc6c205398b61b63d1 --- /dev/null +++ b/checkpoints/028000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:501d004913c6407495bad373133d9a3bc76a928498ca7f5cc8ab05cf84ad95bd +size 2103678952 diff --git a/checkpoints/028000/training_state/rng_state.safetensors b/checkpoints/028000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..6fd5eff1ea3c7c0beacae8af033f7ae12b73bf24 --- /dev/null +++ b/checkpoints/028000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fc4a6505e2255c1ec25676ace72e921dc950e687bcb11c542436fc45a4c96307 +size 15708 diff --git a/checkpoints/028000/training_state/scheduler_state.json b/checkpoints/028000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..dd593f8c61cf0302b647f42f58c7012800a3ede2 --- /dev/null +++ b/checkpoints/028000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 28000, + "_step_count": 28001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 1.1298305091066664e-06 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/checkpoints/028000/training_state/training_step.json b/checkpoints/028000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..86065a62b528a2c5fbe5615c4f34ea67322ce13e --- /dev/null +++ b/checkpoints/028000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 28000 +} \ No newline at end of file